1. What is the rationale (the "why") for combination, triple drug anti-retroviral therapy?
Combination anti-HIV drug therapy provides a powerful inhibition of HIV replication in the body, sometimes dropping viral loads in the blood to "undetectable". The person is not cured, because there are still virus-infected cells in the lymph nodes and possibly other places. But, with the viral load very low, there is greatly reduced killing of T4 cells. In addition, because much less virus replication is occurring, the opportunities for viral evolutionary change may also be significantly lower.
Let's consider an overly simple (but still instructive) hypothetical situation: assume that partial resistance to a specific anti-HIV drug can arise from one specific change in the HIV genome.
Using "order of magnitude" thinking, about 1 in 10,000 virions would have this mutation. Because there are typically about a hundred million or so virions in the body, we conclude that at any given time, there are thousands of virions in the body with the particular change, and thus these virions are already partially resistant to the drug before the patient starts taking it. Because of the high viral turnover rate, these virions will eventually (perhaps quite quickly) take over the population as drug treatment continues.
If the patient starts taking two drugs that require separate mutations to give partial resistance to each drug, the fraction of double-resistant virions already present in the body will be something like 1 in 10,000 times 1 in 10,000, which is 1 in a hundred million. So, there may be at least one virion already partially resistant to both drugs.
If the patient starts taking three drugs that each require separate mutations to give partial resistance to each drug, the fraction of triple-resistant virions already present in the body is expected to be something like 1 in a trillion (one in ten to 12'th power). So, it is extremely unlikely that there is a virion in the body that already has these three mutations.
2. What do we know about HIV transmission?
The single natural process that gives the highest probability of HIV transmission is childbirth by an HIV+ mother who has a high viral load. About 40% to 50% of babies born to high viral load HIV+ mothers get infected either during the pregnancy, during childbirth, or during infancy while breastfeeding. If the woman is on anti-HIV drug therapy, the probability of virus transmission to the fetus/baby is reduced significantly. Even for women not on drug therapy, clinical trials have shown that "A single dose of nevirapine given at the onset of labor plus a single dose to the newborn within 72 hours of birth reduced the risk of HIV transmission down to 13% ." (World Health Organization).
Approximately 10% of worldwide HIV infection totals have been due to mother to child transmission (MTCT). As stated in the link above, "Each year, about 800,000 infants become infected with HIV, mainly through MTCT." In infected infants, the time course until AIDS symptoms appear, and death follows, is almost always much shorter than in adults.
Worldwide, over 70% of HIV infection totals have been due to heterosexual intercourse.
A very interesting "correlate of non-transmission" is the ccr5 genotype of a person. Population genetic analysis has shown that there are two alleles of the ccr5 gene (gene that codes for the CCR5 protein) in the human population; the wild-type allele with an allele frequency of about 90% and a recessive defective allele with an allele frequency of about 10%. Thus, about 1% of the population is homozygous for the defective recessive allele (q=0.1, q squared = 0.01). It turns out that these people are essentially completely resistant to being infected with HIV.
3. What is the story of the first documented case of transmission of a multi-drug resistant strain of HIV from one person to another ?
The first documented case of transmission of multi-drug resistant HIV was published in the New England Journal of Medicine in July 1998, with the title "Sexual transmission of an HIV-1 variant resistant to multiple reverse-transcriptase and protease inhibitors" (Hecht et al.). In 1997, a man ("index patient") was diagnosed as HIV positive just three weeks after a single sexual encounter with a "source patient". The index patient was started on combination drug therapy, but the treatment proved ineffective right from the beginning. A medical history established that the source patient had been on combination anti-HIV drug therapy for over a year but had been "poorly compliant". Virus samples taken from both of these patients were analyzed. A combination of genotype analyses (nucleotide sequencing) and phenotype analyses (in vitro reverse transcriptase and protease enzyme assays in the presence of the drugs) were carried out on both samples and showed that the predominant HIV strain in each person's blood was the same. There were a total of about ten mutations, four in the reverse transcriptase coding region and six in the protease coding region, in this particular multi-drug resistant mutant strain. Compared to wild-type HIV, the relevant enzymes of this mutant strain were about ten-fold more resistant to AZT (zidovudine), another nucleoside analog (lamivudine), and all of the protease inhibitors available.
4. What is an update on HIV evolutionary change and transmission?
See Herbeck et al., "Human Immunodeficiency Virus Type 1 env Evolves toward Ancestral States upon Transmission to a New Host", (Journal of Virology, 2006), and more recent articles that reference this article.
5. What is a 2010 update on development of an effective vaccine against HIV?
Here is the summary of a perspective by Burton and Weiss from Science, 13 August 2010: "A Boost for HIV Vaccine Design".